# ExtraRecursiveReflection

- **Event:** [WeaveHacks 3: Self-Improving Agents Hackathon with Weights & Biases](https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80)
- **When:** Jan 31 at 9:00 AM – Feb 1 at 5:00 PM (PST)
- **Where:** 400 California St, San Francisco, CA
- **Team:** [Darin Kishore](https://cerebralvalley.ai/u/darin)
- **GitHub:** https://github.com/darinkishore/rlm_reflection
- **Demo video:** https://youtu.be/wtfvu1uj2zc
- **Gallery:** https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80/hackathon/gallery/56

Self-improving agents via accumulated reflection. Each iteration: diagnose failure (µf) → minimal skill edit → repeat. 

The key: Recursive LMs operate over the FULL history of attempts, diagnoses, and edits—not stateless reflection. The oracle learns to teach by seeing what worked.

---

Markdown version of https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80/hackathon/gallery/56. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt